1.1 Wetland habitat degradation and loss
Freshwater ecosystems are under severe anthropogenic transformation and over-exploitation worldwide (Dudgeon et al., 2006; Strayer & Dudgeon, 2010; Geist, 2011). The main causes of their deterioration are: exploitation of water and peat resources; wetland conversion to other land use – notably afforestation, agri-culture and urban development (Joosten & Clarke 2002; Silva et al., 2007;
Vörösmarty et al., 2010); external impacts of intensified forestry and agriculture (Williams et al., 2004; Feld et al., 2016; Arntzen et al., 2017; Vilmi et al., 2017);
and fish stock management, drainage and abandonment of cattle ponds (Curado et al., 2011; Lemmens et al., 2013).
The rates of wetland loss have accelerated since the 20th century (Davidson, 2014), and natural wetland area has declined by more than 50% during this period (OECD, 1996; Zedler & Kercher, 2005; Davidson, 2014). Conversion to agricultural land has been the dominant driver. Asia stands out with the largest area of wetlands remaining, but also the largest area lost (Davidson, 2014;
Davidson et al., 2018). Another region with a large historical wetland loss is North America, although the process has somewhat slowed down since the 1980s (Davidson, 2014). Europe has the largest number, but smallest area, of wetland sites among continents; it lost more than half of its natural wetland area before 1990 and the loss has slowed down slightly since then (Acreman et al., 2007;
EEA, 2010; Xu et al., 2019).
In Eastern Europe, including the Baltic countries, wetland loss has been mainly due to draining for agricultural purposes (Hartig et al., 1997). Nowadays vast areas of historical temperate wetlands have shrunk to fragments in pasture areas and agricultural landscapes, losing their functionality as wetland eco-systems (Brinson & Malvárez, 2002). In the Baltic States, Fennoscandia and Russia another major form of wetland exploitation has been artificial drainage for forestry. By the early 1990s, more than 13.5 million hectares of wetlands had been drained for forestry in these regions (Paavilainen & Päivänen, 1995). In addition to inland wetland drainage, formerly grazed wet meadows have decreased on the coasts of the Baltic Sea, e.g. from 29 000 ha to 8000 ha in Estonia; these have turned into scrublands and reed-beds mainly due to disappearance of small farms and cessation of grazing (Luhamaa et al., 2001).
Freshwater ecosystems are long acknowledged to have rich and unique bio-diversity to be sustained (Strayer & Dudgeon, 2010; Moreno-Mateos et al., 2012;
Ramsar, 2018). This is supported by wetland habitat heterogeneity and, often, by relatively high net primary productivity (Tiner, 1984). The human-caused habitat degradation and loss have also led to reduction of wetlands’ biodiversity (Dudgeon et al., 2006; Strayer & Dudgeon, 2010; Geist, 2011). Well known is, for example, the decline of amphibians (Semlitsch, 2002; Stuart et al., 2004; Ficetola et al., 2015; Arntzen et al., 2017) and wetland birds (Wilson et al., 2004; Wetlands
International, 2012; Pearce-Higgins et al., 2017), that are also in the focus of the current doctoral thesis. To reverse these trends, it is critically important to reduce wetland exploitation and restore damaged wetland ecosystems.
1.2 Ecological restoration of wetlands
When disturbance has disrupted ecosystem structure or function beyond an eco-logical threshold, the ecosystem may no longer be able to recover its former state of functioning. This causes also permanent loss of habitats and biodiversity, which can only be restored through specific habitat restoration activities. Such activities are now recognized as a global priority (Aronson & Alexander, 2013). Notably, the 2010 Aichi Targets set an internationally accepted political goal of restoring at least 15% of degraded ecosystems by 2020 (CBD, 2010).
The primary aim of restoration activities is to assist “the recovery of an ecosystem that has been degraded, damaged or destroyed” (SER Primer, 2004).
In the scientific literature, multiple terms and definitions are related to such aim, often interchangeably (e.g. Li, 2006; Lima et al., 2016). For example, van Andel and Aronson (2012) distinguish four main opportunities to restore ecosystems and their constituent habitats, depending on starting conditions and the goal:
near-natural restoration (almost non-assisted natural recovery); true eco-logical restoration (reconstruction of a previous-like state or self-sustaining target) in response to crossed biotic barriers; ecological rehabilitation (improve-ment of ecosystem functions) in response to crossed abiotic barriers; and recla-mation to re-establish productive conditions in heavily degraded lands. Miti-gation is a distinct restoration approach to provide compensation where the impact of disturbance is inevitable (Perrow & Davy, 2002). Another set of terms refers to “creation”, “rehabilitation” and “enhancement”, which are similar to restoration, but differ in some way from the process of renewing natural self-regulating ecosystem (Gwin et al., 1999).
In this thesis I refer to different aspects of habitat restoration as follows:
(re)creating new ponds (i.e., replacement habitats) in paper I; rehabilitation via grazing and mowing on historical coastal meadows in paper II; and enhance-ment via forest partial cutting (III) combined with true ecological restoration of former water regime via ditch blocking (IV) in forested peatlands.
Wetland restoration is a relatively new concept in the history of conservation (Wheeler et al., 1995; Shackelford, 2013), although there has been international attention on conservation and sustainable use of wetlands since the Ramsar Convention (1971). Extensive wetland habitat restoration projects have been carried out in North America (mainly coastal areas), Europe and Asia (Li et al., 2019; Xu et al., 2019). For example, during the last 30–40 years, more than 10 mln ha of North-American wetlands have been restored with variable success (Nadakavukaren, 2011; Copeland, 2017). In Europe, Germany, United Kingdom and France have historically lost large wetland areas (Silva et al., 2007) and now stand out with the largest number of wetland restoration projects (Coops & van
Geest, 2007). In different parts of Europe, there are many successful examples of restoring river floodplain functioning and rewetting polders (Verhoeven, 2014;
EU, 2007), restoring peatlands (e.g. Andersen et al., 2017; Brown et al., 2016;
Menberu et al., 2016), wet grasslands (EU, 2007; Joyce, 2014) and ponds (Rannap et al., 2009b). Mainly in Asia, a significant share of all wetlands is artificial – created by converting natural wetlands into rice paddies (Leadley et al., 2014).
Due to agricultural intensification, these wetlands are losing their values as com-pensating areas for ecosystem functions and biodiversity (Katayama et al., 2015;
Giuliano & Bogliani, 2019). Restoration of wetland ecosystems can provide various important services, including a major role in carbon accumulation; water quality, storage and regulation; nutrient cycling; habitat provisioning for aquatic and semi-aquatic biodiversity; cultural heritage and recreation for people (Frolking et al., 2006; Kimmel & Mander, 2010; Lamers et al., 2015).
Despite long term restoration efforts, there are many obstacles to sufficient conservation of wetlands, particularly in densely populated areas of Asia (Kentula, 2000; Moreno-Mateos et al., 2012; Prigent et al., 2012; Choi, 2004). The obstacles include: a lack of scientific understanding of wetlands complexity and causal path-ways to modify their ecosystems and assemblages; unclear objectives and criteria of restoration success; and multiple social, economic, and political constraints.
An important principle of successful restoration is that each ecosystem is approached individually and contextually (Kovalenko et al., 2012; Tokeshi &
Arakaki, 2012). However, specifically for the purpose of preserving biodiversity, it is difficult and laborious to detect, monitor or manage every aspect of species and habitat. Thus, shortcuts are sought to reasonably simplify the management and speed up knowledge acquisition. Among such shortcuts, monitoring and managing for a few carefully selected species (termed, e.g., focal species, umbrella species, flagship species, target species) has long been one of the key issues in con-servation biology (Simberloff, 1998; Caro & O’Doherty, 1999; Caro, 2010).
1.3 The focal species approach
This dissertation addresses the concept of restoring and managing habitats and ecosystems according to selected specialized ‘focal’ taxa. According to Lambeck (1997) focal species are defined as the most sensitive species to individual threats in a changing environment, representing four main categories: area-, resource-, process- and dispersal-limited species. So far, the practical use of focal species approach (often included as set of umbrella species that indirectly protect many other species; Roberge & Angelstam, 2004) has been mainly confined to a (virtual) selection of strict protected areas (Rodrigues & Brooks, 2007; Seddon
& Leech, 2008; Caro, 2010). However, its application to habitat management, (e.g. Simberloff, 1998; Caro & O’Doherty, 1999; Carignan & Villard, 2002;
Roberge & Angelstam, 2004) and restoration (e.g., Petranka & Holbrook, 2006;
Kumar et al., 2018) has been also debated worldwide.
Birds and environmentally sensitive aquatic animals are among the most attractive groups of focal species, since they are relatively easy to sample and possess unique habitat requirements. Specifically, for wetland conservation, they are intimately connected with the hydrologic conditions of ecosystems (Suter et al., 2002; Balcombe et al., 2005; Caro, 2010).
Although selected bird species may not be the most appropriate indicators of species richness of other taxon groups (Lund & Rahbek, 2002; Xu et al., 2008), they can effectively indicate full avian biodiversity of conservation interest (Suter et al., 2002; Senzaki & Yamura, 2016). Due to birds’ sensitivity to anthropogenic perturbations (Brawn et al., 2001), they have practical value for prioritization of larger areas for conservation planning (Roberge & Angelstam, 2004; Alexander et al., 2017) and for habitat management (Paillisson et al., 2002). Also, birds can indicate certain habitat qualities (Rempel et al., 2016; Vallecillo et al., 2016) and aspects of restoration success (Crozier & Gaulik, 2003).
The evidence of amphibians as focal species in freshwater ecosystems is contradictory. Across landscapes that contain both terrestrial and aquatic habitat, amphibians are probably poor cross-taxon indicators (Beazley & Cardinal, 2004;
Xu et al., 2008; Ruhi et al., 2014; Vehkaoja & Nummi, 2015). However, because of limited dispersal abilities (Smith & Green, 2005; Kovar et al., 2009) and high site fidelity (e.g. Loman, 1994; Bucciarelli et al., 2016), amphibians may have indicator value for conservation purposes on a more local scale. Amphibians might also indicate the success of ecosystem restoration (e.g. Waddle, 2006;
Welsh & Hodgson, 2013; Diaz-Garcia, 2017), specifically in aquatic habitat restoration (Price et al., 2007). For example, in Italy, Rana italica has been proposed as a bioindicator for water quality condition in small headwater streams (Lebboroni et al., 2006). Welsh and Ollivier (1998) showed amphibian suitability as indicators of stream ecosystem dysfunction after road construction and fine sediment pollution into pristine streams. Creation and restoration of temporary waterbodies have also been assessed based on amphibians’ response, in com-parison with natural and restored/created water bodies (Kolozsvary & Holgerson, 2016; Rothenberger et al., 2019).
In this dissertation, I used meadow birds and pond-breeding amphibians to examine the effects of habitat management in three different wetland systems.
Among birds, I selected Baltic dunlin (Calidris alpina schinzii) to study the habitat change and overall assemblage richness along with management intensity (mowing, grazing) of restored coastal meadows (II). Among amphibians, I explored great crested newt (Triturus cristatus) and common spadefoot toad (Pelobates fuscus) in relation to pond creation (comparing natural, man-made and specifically constructed ponds) and suitability for broader amphibian and macro-invertebrate assemblages (I). I also used more widely distributed brown frogs – moor frog (Rana arvalis) and common frog (R. temporaria) – to evaluate habitat changes in the aquatic habitats of degraded peatland forest ecosystems after their ecological restoration for a protected bird, western capercaillie (Tetrao urogallus) (III, IV).
1.4 Aims and motivation
The general aim of my studies was to explore wetland management and resto-ration effectiveness, notably for threatened wetland assemblages, as guided by habitat-sensitive focal species. The dissertation consists of four case studies in different wetland realms in Estonia: ponds (I), managed wet grasslands (II), and peatlands drained for forestry (III, IV).
I use focal taxa to address three broader knowledge gaps in the management of wetland habitats through focal species. First, the relationship between focal taxa and assemblage structure (I). I studied small water bodies within terrestrial habitat mosaics, which have been degraded due to intensive agriculture, forestry drainage and abandonment of traditionally managed lands (Céréghino et al., 2008; Curado et al., 2011). As a restoration opportunity, threatened and wide-spread pond-breeding amphibians readily inhabit newly created or reconstructed ponds (Rannap et al., 2009b; Magnus & Rannap, 2019) but it is not known whether the assemblages are linked enough to use the amphibian targets also for other co-occurring (semi) aquatic taxa in such restoration. Formally, then, confirming assemblage nestedness might be an effective step in focal species selection (Beazley & Cardinal, 2004).
Secondly, relationships between the abundance of waders and broader diver-sity of coastal grasslands (II). Managed wet coastal grasslands are known to sup-port diverse plant communities, provide breeding grounds for amphibians and threatened birds (Paal, 1998; Kuresoo et al., 2004; Rannap et al., 2007). After being abandoned as traditional agricultural areas, the diversity declines, often despite using modern approaches to conservation management. This is probably due to insufficient knowledge of exact habitat characteristics that have histori-cally supported viable populations of coastal-meadow species.
Thirdly, frog breeding in drained forested peatlands as an indicator for habitat management and restoration options. The drainage ditches, which substitute natural depressions in these ecosystems, are considered to be attractive (Remm et al., 2015) but low-quality breeding sites for amphibians (Suislepp et al., 2011). I studied the impact of forest partial cut (III) and ditch blocking manipulations (IV) on brown frogs breeding habitat, and the most preferred target conditions.
I address the following study questions:
i. Does habitat management for threatened species create quality habitats also for other species of conservation concern (as compared to non-managed areas), i.e., is increased abundance of focal species accompanied by a generally higher diversity of wetland-dependent species (I, II)?
ii. Are the key factors that shape assemblages similar in all three wetland study systems (I–IV)? Which habitat characteristics are most important for the nesting of threatened focal species on coastal meadows (II)?
iii. To what extent do reduced shade and changed water regime mitigate the drainage effects in forest areas where only ditches have remained as breeding sites for amphibians? Does ditch blocking favour the formation
of other types of small water bodies? How rapidly do frogs colonize the improved breeding habitats (I, III–IV)?
iv. What are the opportunities of using selected wader and amphibian species as focal species for conservation management and restoration (I–IV)? Is nestedness analysis a useful tool for the practical task of selecting focal species for habitat conservation (I)?